Over 5 million Americans use marijuana at least once a week and approximately 3.1 million use marijuana daily or almost daily. Despite the relatively high prevalence of use in the population, little is known about the long term health risks of marijuana. Two studies are proposed by supplementing the data scheduled to be collected on the 12 year follow-up of the Baltimore Epidemiologic Catchment Area (ECA) Study sample in 1993. These two studies will evaluate marijuana use as a potential risk factor for cancer and evaluate the comprehensive health status of individuals who use marijuana use. Study 1 is a case-control study to evaluate marijuana use as a risk factor for head, neck and lung cancer in adults fifty years or less. Incident cases (n=270) will originate from the Baltimore metropolitan area and controls will be from the community-based ECA study respondents. Both cases and controls will be interviewed concerning lifetime and current marijuana use as well as potential confounders such as tobacco exposure, alcohol use, diet, occupation, and psychiatric status. No explicit matching of cases and controls will occur. Estimates of the association of marijuana use and head, neck and lung cancer with adjustment of potential confounders will be calculated using logistic regression. Study 2 is a cross-sectional and longitudinal study of the entire Baltimore ECA sample less than fifty years of age (n-1200). Comparisons will be made of current users, quitters and never users of marijuana in terms of their health status, health behaviors, health risks, and use of health, particularly preventive, services. Results of this study will provide new knowledge of how marijuana use impacts on a broad array of health measures. This information should be of benefit to future investigators attempting to disentangle the medical consequences of drug abuse as well as provide information to guide health professionals attempting to design programs to reduce the health risks of illicit drug users.